AgeCommit message (Expand)AuthorFilesLines
2019-09-11Bump version for rc3mesa-19.2.0-rc3Dylan Baker1-1/+1
2019-09-10radeonsi/gfx10: fix wave occupancy computationsMarek Olšák4-21/+49
2019-09-10radeonsi/gfx10: don't call gfx10_destroy_query with compute-only contextsMarek Olšák1-1/+1
2019-09-10virgl: Fix pipe_resource leaks under multi-sample.Lepton Wu1-1/+3
2019-09-09st/nine: Properly initialize GLSL types for NIR shaders.Timur Kristóf1-0/+5
2019-09-09Revert "ac/nir: Lower large indirect variables to scratch"Bas Nieuwenhuizen1-68/+0
2019-09-09nir/dead_cf: Repair SSA if the pass makes progressJason Ekstrand1-2/+13
2019-09-09nir/repair_ssa: Insert deref casts when neededJason Ekstrand1-2/+29
2019-09-09nir/repair_ssa: Repair dominance for unreachable blocksJason Ekstrand1-4/+8
2019-09-09nir: Add a block_is_unreachable helperJason Ekstrand2-0/+15
2019-09-09nir: Don't infinitely recurse in lower_ssa_defs_to_regs_blockJason Ekstrand1-5/+15
2019-09-09nir: Handle complex derefs in nir_split_array_varsJason Ekstrand1-2/+5
2019-09-09drirc: override minImageCount=2 for gfxbenchEric Engestrom1-0/+4
2019-09-09radv: add support for vk_x11_override_min_image_countEric Engestrom1-0/+1
2019-09-09amd: move adaptive sync to performance section, as it is defined in xmlpoolEric Engestrom2-5/+2
2019-09-09anv: add support for vk_x11_override_min_image_countEric Engestrom1-0/+3
2019-09-09wsi: add minImageCount overrideEric Engestrom5-3/+27
2019-09-09anv: add support for driconfEric Engestrom4-3/+19
2019-09-09anv: Bump maxComputeWorkgroupSizeJason Ekstrand1-4/+6
2019-09-06nir/lower_explicit_io: Handle 1 bit loads and storesCaio Marcelo de Oliveira Filho1-9/+24
2019-09-06freedreno/a2xx: ir2: fix lowering of instructions after float loweringJonathan Marek1-3/+2
2019-09-06intel/dri: finish proper glthreadSergii Romantsov1-1/+1
2019-09-05broadcom/v3d: Allow importing linear BOs with arbitrary offset/stride.Dave Stevenson1-8/+23
2019-09-05radv: Call nir_propagate_invariant()Connor Abbott1-0/+2
2019-09-04nir: do not assume that the result of fexp2(a) is always an integralSamuel Pitoiset1-0/+1
2019-09-04nir: Add is_not_negative helper functionDylan Baker1-0/+6
2019-09-04glx: Fix SEGV due to dereferencing a NULL ptr from XCB-GLX.Hal Gentz2-1/+9
2019-09-04iris: Report correct number of planes for planar imagesKenneth Graunke1-1/+8
2019-09-04nir: fix memleak in error pathEric Engestrom1-1/+3
2019-09-04freedreno/drm-shim: fix mem leakEric Engestrom1-3/+4
2019-09-04anv: fix format string in error messageEric Engestrom1-1/+1
2019-09-04util/os_file: fix double-close()Eric Engestrom1-1/+0
2019-09-04egl: fix deadlock in malloc error pathEric Engestrom1-1/+3
2019-09-04ttn: fix 64-bit shift on 32-bit `1`Eric Engestrom1-1/+1
2019-09-04vulkan/overlay: bounce image back to present layoutLionel Landwerlin1-2/+30
2019-09-04egl: fix platform selectionLionel Landwerlin1-2/+7
2019-09-04gallium/auxiliary/indices: consistently apply start only to inputErik Faye-Lund1-10/+10
2019-09-04travis: Fail build if any command in if statement fails.Vinson Lee1-4/+4
2019-09-04swr: Fix build with llvm-9.0 again.Vinson Lee3-0/+28
2019-09-04bump version to 19.2-rc2mesa-19.2.0-rc2Dylan Baker1-1/+1
2019-09-04glsl: replace 'x + (-x)' with constant 0Pierre-Eric Pelloux-Prayer1-0/+12
2019-09-04Revert "radeonsi: don't emit PKT3_CONTEXT_CONTROL on amdgpu"Thong Thai1-7/+4
2019-09-04nir/range-analysis: Handle constants in nir_op_mov just like nir_op_bcselIan Romanick1-2/+9
2019-09-04nir/range-analysis: Fix incorrect fadd range result for (ne_zero, ne_zero)Ian Romanick1-3/+8
2019-09-04nir/range-analysis: Adjust result range of multiplication to account for flus...Ian Romanick1-31/+22
2019-09-04nir/range-analysis: Adjust result range of exp2 to account for flush-to-zeroIan Romanick1-2/+14
2019-09-04nir/algebraic: Mark some value range analysis-based optimizations impreciseIan Romanick1-9/+13
2019-09-04iris: Fix partial fast clear checks to account for miplevel.Kenneth Graunke1-2/+2
2019-09-04intel/compiler: Request bitfield_reverse lowering on pre-Gen7 hardwareIan Romanick1-0/+1
2019-09-04nir/algrbraic: Don't optimize open-coded bitfield reverse when lowering is en...Ian Romanick1-1/+1